The North Atlantic Treaty Organization (NATO) is in crisis. Just over a year after Lloyd Austin, then the US defence secretary, called it ‘the most powerful and successful alliance in history’, a new administration in Washington has cast doubt on America’s commitment to European defence. Although officials and observers were given ample warning signs from the new president, Donald Trump, there is nonetheless a pervasive sense of astonishment and dismay among European leaders over NATO’s rapid and possibly imminent demise.

Stepping back, their sense of shock is made more understandable. Policymakers and international scholars alike have had their conceptions of alliances shaped primarily by the dominant schools of neorealist or neoliberal institutionalist thought. The former (focusing on the crucial role of state-versus-state power calculations) and the latter (emphasising the cumulative benefits of bureaucratised cooperation) have prepared us poorly to anticipate and address alliance dynamics.

Perhaps if leaders had spent a bit more time absorbing the wisdom of classical realists in international relations — scholars such as George Kennan, Hans Morgenthau, and especially Reinhold Niebuhr — they would find themselves better prepared for this historic moment. While traditional realists differed much about world affairs and foreign policy, they agreed on fundamental truths such as the centrality of states in the international system, the premium that leaders put on survival and security, and the danger of an idealistic faith in human nature.

Most notably, however, these titans of 20th-century global politics used their wisdom and experience — in Kennan’s case, by actually contributing to the formulation and execution of US foreign policy — as well as their deep respect for the richness of historical and sociological insights in illuminating our understanding of international affairs.

Through their recognition of the profound effects of shared identity and responsible leadership on alliance durability, classical realists warned us that rapid changes to either one would probably undermine even a well-established alliance, such as NATO, by reducing the pact’s effectiveness and even risking its dissolution. Today, the Trump administration’s sudden downgrading of Europe in its strategic calculus clearly and directly affects both of those factors.

Neorealists, such as Kenneth Waltz in his groundbreaking Theory of International Politics, assert that focusing on the balance of power between states offers the most bang for the buck when analysing international relations, including alliance dynamics. Joseph Grieco supplements Waltz by highlighting that even mutually beneficial cooperation among nations allows someone to achieve relative gains — inhibiting durable commitments because ‘today’s friend may be tomorrow’s enemy in war’. All things being equal, adherents of this view assert that international alliances are brittle: they rely on constantly shifting calculations of relative power and are subject to recalculation and realignment when shocks to the systemic balance of power occur.

Conversely, neoliberals and their idealist intellectual cousins see international organisations as inherently durable once cooperative institutions begin to take hold. Alliances may have a unique core function among international institutions of enhancing security, but, just like other such organisations, alliances over time reduce the costs of future cooperation, raise the costs of cheating, and enmesh members within increasingly complex webs of formal and informal interactions. Proponents of this view argue that the more institutionalised the alliance is, the more likely it is to endure the type of shocks that neorealism would generally consider as fatal.

Both frameworks have useful dimensions. Neorealists don’t let us forget that power remains central to international politics and that groupings beyond the nation state must overcome tremendous challenges to survive within an anarchic global environment. Institutionalists remind us that communities and commitments between states develop into networks and norms that also have power.

Neither school of thought, however, serves as a reliable guide to the resilience of alliances, because neither one captures as well as classical realism the fundamental truths about such pacts. These truths can be summarised as follows: shared identity makes alliances sticky, and enlightened leadership extends their life. When an alliance member, especially a dominant one, threatens either one of those, even a longstanding alliance is at risk of falling apart.

Central to the wisdom that classical realism provides about alliance dynamics is the respect that its deepest thinkers had, not only for the centrality of power — going back at least to Thucydides — but also for the dynamics of inter-group relations. Identities, not just at the level of the nation-state, are crucial. Robert Gilpin identifies three key realist assumptions, one of which is that ‘the essence of social reality is the group… This is another way of saying that in a world of scarce resources and conflict over the distribution of those resources, human beings confront one another ultimately as members of groups, and not as isolated individuals. Homo sapiens is a tribal species, and loyalty to the tribe for most of us ranks above all loyalties other than that of the family’.

The writings of Reinhold Niebuhr, whom fellow realist George Kennan honoured as ‘the father of all of us’, also capture the group-based nature of politics: ‘The individual cannot be a true self in isolation. Nor can he live within the confines of the community which “nature” establishes in the minimal cohesion of family and herd.’ He observes that ‘Men become conscious of themselves as they see themselves in relation to other life and to their environment.’

The pessimism often associated with classical realist thought derives from its conservative foundation: human nature is flawed, making intergroup conflict fundamental. Niebuhr writes in Moral Man and Immoral Society (1932):

The increasing size of the group increases the difficulties of achieving a common group purpose, except as it comes into conflict with other groups and is unified by perils and passions of war. It is a rather pathetic aspect of human social life that conflict is a seemingly unavoidable prerequisite of group solidarity.

A few years later, the influential Hans Morgenthau encapsulated this basis for the traditional realist understanding of world politics: ‘Tensions are a universal phenomenon of social life.’ Idealists’ efforts to alleviate the negative aspects of human nature and the ubiquity of intergroup conflict via a world state fail because, as Niebuhr points out, ‘the community of mankind … is too vague to inspire devotion’.

Somewhere in between the family and all humanity, people find a way to gather in political units of different sizes and shapes. Over the long history of regional and world politics, nation states are but recent creations; they have not been and need not now be the only group on the global stage eliciting identification and commanding loyalty. While the state, for all realists, is the default group identity in international relations due to its ubiquitous influence in world affairs, other groups — including alliances — can become the object of intense identification in certain contexts. Again in Moral Man and Immoral Society, while Niebuhr notes that ‘the increasing size of the group increases the difficulties of achieving a group self-consciousness’, he nevertheless leaves the door open for the importance of larger groups by adding, ‘except as it comes into conflict with other groups and is unified by parallels and passions of war’.

Gilpin writes:

In the modern world, we have given the name “nation-state” to these competing tribes and the name “nationalism” to this form of loyalty. True, the name, size, and organization of the competing groups into which our species subdivides itself do alter over time — tribes, city-states, kingdoms, empires, and nation-states — due to economic, demographic, and technological changes. Regrettably, however, the essential nature of intergroup conflict does not.

Separately, he notes:

The fundamental unit of social and political affairs is the group, or what the distinguished German sociologist Ralf Dahrendorf has called the “conflict group”. The precise nature of these conflict groups, however, has changed over the millennia in response to economic, technological, and other developments. In Aristotle’s Greece, for example, the basic unit was the polis or city-state. In the modern world, the principal conflict group has been the territorial state whose foremost manifestation today is the nation-state.

Kennan, in his 1951 book, American Diplomacy, put it succinctly: ‘The national state pattern is not, should not be, and cannot be a fixed static thing.’

Among the classical realists, Niebuhr most explicitly remains open to the variety and progression of political collectives, including those beyond the nation state:

The variations in the pattern of international community vary endlessly. But the fixed pattern in all of these variations is a combination of the dominion and authority above the level of the nation and below the level of the community of mankind. It is safe to predict that no future history will annul this pattern, though it may produce hitherto unknown variations in the pattern.

A classical realist explanation of alliance durability starts therefore from the same baseline as the neorealist perspective: alliances are almost entirely functions of the international structure of power, forming and dissolving according to forces in the environment. Classical realists, however, understand that alliance members may see each other to varying degrees as members of the same in-group, especially under conditions of severe and prolonged threat.

In-group identification may indeed be more difficult to achieve above the country level, but the sense of shared identity that occasionally arises within such groupings contributes to alliance durability. Most of the time, especially when facing less than lethal threats, alliances fail to elicit identification nearly as deep as that elicited by states. When the pressure from some ‘other’ drives more intense perceptions of the alliance as an in-group, however, this shared identification will have an important lag effect. Again, in Niebuhr’s words, ‘Any distinguishing mark between the “we group”, in which mutual responsibilities are acknowledged, and a “they group”, who are outside the pale of humanity, may serve the tribal character of human nature’.

This explains why even weakly institutionalised alliances may endure shocks that should, from neorealist and neoliberal institutionalist points of view, break them up. Kennan perhaps puts it best, writing that intense, prolonged alliances between nations ‘have a habit of surviving the situations that gave rise to them in the first place. Their very existence creates new situations, which their creators could never have envisaged’.

Put simply, neoclassical realism suggests that an alliance is more likely to endure shocks if its member states have experienced the growth of a sense of shared identity as an in-group. A powerful and aggressive outsider that is perceived as a common threat can work wonders to solidify what may have previously been very weak links of common identification. During the Cold War, the varying but never absent threat across several decades from Moscow, for example, helped NATO survive internal conflicts until the fall of the Soviet Union 35 years ago, and then remain together until the threat from Russia resurged in the past decade.

Brendan Simms highlights that treaties rely on trust — which, he notes, ‘takes a long time to build’. But, giving a nod to the relationship between deep trust and a sense of shared identity, he adds, ‘it is not easily lost’.

One way to lose it relatively quickly is regime change in a prominent alliance member, which can undermine an alliance’s sense of shared identity. Classical realism provides the logic for George Liska’s observation that the ‘most straightforward cause of the disintegration of an alliance is domestic instability, producing radical change in the governing elite’, because such changes drastically alter perceptions of in-group identification.

Separately, classical realists also understand the value of responsible alliance leadership. Whereas neorealists posit that all states, even those in seemingly secure alliances, must look at each other as potential opponents due to the implications of anarchy, classical realists don’t let their acknowledgement of such tensions blind them to the crucial factor of enlightened leadership in bolstering alliance solidarity.

Kennan notes the ‘reasonable level of dignity and deference’ leaders should be able to expect. ‘Where leadership is quietly and unostentatiously exercised’, he asserts, ‘with due respect for those over whom it is exercised, and with a sense of responsibility no smaller than the authority it implies, no exception can be taken to it.’

According to Gilpin, along with many other thinkers within and without the realist tradition, America’s role in solidifying the West and its liberal trading order after the Second World War showed such responsible leadership. The United States, he observes, ‘eschewed the temptation to exercise its political and economic power for nationalistic ends’. This idea is extended to alliance durability by Stephen Walt, who writes that ‘alliances are more likely to persist when a strong power is both willing and able to pay the cost associated with leadership’.

In his 1944 work, The Children of Light and The Children of Darkness, Niebuhr chastises interwar idealists, whom he claims, ‘failed to understand the persistence and power of the pride of nations’. He warns that most countries are simply ‘too proud to understand that their power might be a peril to other nations’.

Enlightened leaders recognise the negative consequences of such pride and seek instead to bear the disproportionate costs of alliance leadership and abjure opportunities to use their dominant position to take all they can get from their international partners.

Small states may ally with a strong state both for the immediate security benefits from capability aggregation and for the status and prestige benefits to the governing regime from such an alliance. Irresponsible leadership, however, undermines both motivations and ensures that any resulting alliance will all too easily fall prey to the disintegrative effects of the international system.

Niebuhr, true to form, is highly perceptive on this subject:

If America achieves maturity, the primary mark of it must be the willingness to assume continuing responsibility in the world community of nations. We must seek to maintain a critical attitude toward our own power impulses; and our self-criticism must be informed by the humble realization of the fact that the possession of great power is a temptation to injustice for any nation… The forces of self-interest to be deflected are not always those of the opponent or competitor. They are frequently those of the self, individual or collective.

Deeply aware of history, classical realists pull these lessons about alliance leadership from centuries and millennia past — back to the Delian League in ancient Greece. When Athens began behaving less like a responsible leader and more like an imperial overlord, the cohesion from Persian invasions dissipated and the alliance’s days were numbered. Henry Kissinger put his deep sense of history into practice as US secretary of state and national security advisor; T.G. Otte notes Kissinger’s understanding that ‘the exercise of power implies an obligation to accept the limits of self-restraint’, that being overbearing carries as much danger as abdicating leadership.

Thus, classical realism suggests that an alliance is more likely to endure shocks if it has a dominant leader exercising its primary role with restraint and prudence. Responsible leaders do not use their relative strength to humiliate their partners, and the alliances they lead will be more durable as a result of such prudence.

No international grouping lasts forever. Alliances in particular, subject to the ebbs and flows of relative national power, form and break and form again.

NATO nevertheless has been a remarkably durable institution, able to withstand shocks like the Suez Crisis in the 1950s, the French withdrawal from the alliance’s integrated military command in the 1960s, disputes over nuclear deployments in the 1970s and 1980s, the dissolution of the Soviet Union in the 1990s, and enlargement debates in that decade and ever since. It may not be done yet.

However, the slow but steady movement since 1947 towards a shared identity and the generally responsible alliance leadership across many decades by the United States looks highly questionable as the new administration in Washington conducts unorthodox — and, to longstanding security partners, bewildering — policies, such as threatening allies. This shock, coming so dramatically from within the alliance, in a way unseen in its more than 75-year history, hits right at what Niebuhr, Kennan, and others wisely identified as the very factors allowing beneficial alliances to endure in an anarchic system.

It seems appropriate to close with Niebuhr’s warning in Christianity and Power Politics (1940), as crucially relevant to some leaders now as it was to the isolationists of his times:

The isolationism of America belongs in the same category of political facts as the complacency of the British people during the Munich crisis. In both cases, the general public did not understand strategy well enough to know that by yielding to a tyranny now, or by sacrificing allies and refusing them help, it was merely hastening the day when it would have to face that same tyranny with fewer resources.