The Rotating Door of Nations: What Five Thousand Years of Abandoned Alliances Teaches America About Its Friends

There is a phrase that recurs, with minor variations, in the diplomatic correspondence of virtually every major power across recorded history: the situation has changed. It appears in Hittite clay tablets explaining the recalibration of treaty obligations. It appears in the dispatches of Renaissance Italian city-states pivoting between French and Spanish patronage. It appears in the cables of twentieth-century foreign ministries navigating the shifting geometries of the Cold War. And it appears, in various formulations, in the contemporary language of American foreign policy.

The situation has changed. And with it, the alliance.

The record of man on this subject is unambiguous to the point of discomfort: international commitments endure precisely as long as the strategic calculation that produced them. When that calculation changes — as it inevitably does — the commitment is renegotiated, qualified, or abandoned, regardless of what the treaty language says, regardless of what was promised, and regardless of what the abandoned party was led to believe.

The Ancient Pattern

The diplomatic archive of the ancient Near East, particularly the Amarna letters of the fourteenth century BC, provides the earliest extensive record of alliance politics, and it reads with a familiarity that is almost jarring. The letters exchanged between the Egyptian pharaoh and the kings of Babylon, Assyria, Mitanni, and the Hittite empire are saturated with complaints about broken promises, unreturned gifts, and the perception of being treated as lesser partners than the relationship warranted.

King Tushratta of Mitanni, writing to Pharaoh Akhenaten, spends considerable diplomatic energy lamenting that the gold statues promised by Akhenaten's father Amenhotep III had been replaced with gold-plated wooden substitutes — a detail that would be comic if it did not so precisely capture the dynamic of a great power managing a secondary partner's expectations downward as the strategic value of the relationship declined.

Mitanni was, within a generation of these letters, effectively erased as a geopolitical entity — caught between the expanding power of the Hittites and the Assyrians, and unable to secure the Egyptian military support that the alliance had, in theory, guaranteed. The situation had changed. Egypt had its own concerns. Mitanni had served its purpose.

The Mechanics of Alliance Abandonment

Understanding why alliances fail requires distinguishing between the stated basis of an alliance and its actual basis. The stated basis is almost always a combination of shared values, mutual defense commitments, and the language of partnership. The actual basis is almost always a convergence of strategic interests that may or may not have anything to do with values, and that exists only as long as the convergence persists.

This distinction is not cynical. It is descriptive. The historical record shows that alliances framed in the language of shared values are not more durable than alliances framed in purely transactional terms — they are simply more likely to produce a sense of betrayal when the underlying strategic interests diverge, because the abandoned party believed the values were load-bearing when they were, in fact, decorative.

The United States has been on both sides of this dynamic with sufficient frequency that the pattern should, by now, be familiar.

The American alliance with the Shah of Iran, maintained through the 1950s, 1960s, and 1970s on the basis of anticommunist strategic alignment, was framed in the language of modernization, partnership, and regional stability. When the Shah fell in 1979, the United States discovered that decades of relationship had produced an Iranian population that experienced the alliance primarily as support for an authoritarian government, and a successor regime that treated American friendship as evidence of prior complicity.

The strategic interest that had sustained the alliance was real. The values framing that surrounded it was not sufficient to survive the strategic interest's disappearance.

The Forgotten Friends

Perhaps the most consistent feature of alliance abandonment across the historical record is not the abandonment itself — that is, as established, structurally predictable — but the genuine surprise of the party being abandoned. This surprise is itself historically consistent, and it reveals something important about how states process the difference between what alliances say and what they do.

The South Vietnamese government in 1975, the Kurdish fighters who served as American proxies in multiple theaters across multiple decades, the Afghan military and civilian infrastructure built at extraordinary cost over twenty years — each of these cases produced the same spectacle of an abandoned partner expressing shock that the commitment had proved conditional. Each time, American officials and commentators responded with some variation of the observation that the situation had changed, that the commitment had always been contingent, that the partner had been advised of the limits.

The historical record suggests that this exchange — the abandoned party's surprise, the abandoning party's insistence that the limits were always clear — is itself a ritual, not a genuine communication failure. The abandoned party knew, at some level, that the commitment was contingent. The abandoning party knew, at some level, that it had encouraged the belief that the commitment was more absolute than it was. The surprise is performed, at least in part, because the alternative — acknowledging the transactional reality of the relationship throughout — would have made the alliance harder to sustain while it was useful.

The Question of American Reliability

For an American audience, the historical record on alliance reliability poses a question that is simultaneously strategic and moral: what does the pattern of alliance abandonment cost the United States in terms of its ability to form useful partnerships in the future?

The historical evidence on this question is mixed in ways that are worth examining carefully. On one hand, the record shows that states seeking allies are generally pragmatic enough to work with partners whose reliability is known to be conditional — because, in the final analysis, all partners' reliability is conditional. France and the United States have a relationship defined by periodic, sometimes spectacular mutual betrayal, and they remain closely aligned because the underlying strategic interests continue to converge.

On the other hand, the record also shows that the reputational cost of alliance abandonment is not zero, and that it accumulates in specific contexts. Partners who lack the strategic weight to be useful to a major power regardless of their reliability — smaller states, non-state actors, regional partners in peripheral theaters — factor the major power's abandonment history into their calculations in ways that affect what they are willing to commit, what intelligence they are willing to share, and how far they are willing to go in support of operations that require their full participation.

In other words, the United States pays for its alliance abandonment history not in relationships with France or Germany or Japan — states large enough that the relationship is self-sustaining on strategic grounds — but in the accumulated hesitation of every smaller partner that has watched what happened to the last smaller partner.

What the Record Suggests

Five thousand years of diplomatic history does not yield a clean prescription for American foreign policy. What it yields is a set of observations that are useful precisely because they are unsentimental.

Alliances formed around genuine, durable strategic interests tend to outlast alliances formed around values alone. Commitments that are honestly communicated as conditional tend to produce less catastrophic ruptures than commitments presented as unconditional. And the reputational cost of abandonment, while not fatal to major-power relationships, is real and cumulative in the relationships that most require trust to function.

The situation will always change. The question is whether American foreign policy is capable of building alliances that are honest about that fact from the beginning — and whether any administration has yet found the political language to tell a prospective partner, clearly and at the outset, that the friendship is real, and conditional, and will last exactly as long as it needs to.

The record suggests that no one has quite managed it. But the record also suggests that the attempt would be worth making.