Why a Superpower Picks Fights With Its Own Backyard

The situation

The United States has spent several years pressuring its two neighbours: tariffs on Canada, talk of making it the "51st state," threats of strikes on Mexican soil, economic coercion aimed at both. The stated reasons shift — trade deficits, fentanyl, migration. Step back and a stranger question appears: why would the strongest country on Earth strain the two relationships that quietly underwrite its strength?

The model: the stopping power of water, and the secure home region

Geopolitics studies how a country's position on the map shapes its power. One of its oldest findings: a state's security depends heavily on its neighbourhood. The naval strategist Alfred Thayer Mahan, and later the geopolitical theorist Nicholas Spykman in the 1940s, argued that the US enjoyed a near-unique gift — weak, peaceful neighbours and two ocean moats — which freed it to project power abroad instead of defending its own ground. The political scientist John Mearsheimer later formalised the ocean half of this as "the stopping power of water": large bodies of water make invasion of a homeland extraordinarily hard, so a power protected by them can act as the dominant player in its own hemisphere and intervene selectively elsewhere.

The core variable is neighbours. Every border is a relationship to manage. Friendly neighbours are a subsidy: they let you ignore your flanks. Hostile ones are a tax: they force you to garrison, worry and spend. The asymmetry among major powers is stark:

China: 14 land neighbours, several hostile, sea access running through chokepoints others can close.
Russia: 14 neighbours, a long history of war with many.
India: 6 neighbours, including a nuclear-armed rival it has fought repeatedly.
United States: 2 neighbours, both peaceful, open ocean on two sides, no serious threat to the homeland.

This is America's deepest advantage — deeper than its economy or military, because those were built on top of it. And here is the point the current fight ignores: this advantage is not a fact of geography. It is a relationship. The water is permanent. Canada and Mexico are not inert terrain; they are states that choose, day by day, whether to be a subsidy or a tax.

The mechanism: how a quiet border becomes loud

Reagan put the strategy plainly in 1979: "The key to our own future security may lie in both Canada and Mexico becoming much stronger than they are today." A benign neighbour does specific work:

Lowers defence costs — you thin out home forces.
Builds a shared market of resources, talent and supply chains at a scale no rival continent can match.
Handles transnational problems at the source — crime, migration, smuggling — because it cooperates rather than exports the trouble.
Presents a common front, so Washington bargains with the world backed by three economies, not one.

Coercion flips the incentive behind each step. When a neighbour comes to see you as a threat to its economy and independence, cooperation on crime and migration becomes a chip to withhold, not a shared project. It diversifies trade away from you — toward China, toward Europe — to cut its exposure. It stops presenting a common front, because you are now part of what it defends against.

The US does not lose its geography. It loses the relationship that converts geography into power. The moat remains; the subsidy is gone. In Mexico, the pressure revives a century-old fear of "gringo hegemony" that makes every future ask costlier.

History shows the bill directly. When Washington took continental peace for granted in the early twentieth century, the Mexican Revolution turned the southern border into a source of danger. In 1917 Germany secretly offered Mexico an alliance to recover Texas, Arizona and New Mexico — betting an unstable neighbour could become a dagger at America's back. A troubled neighbour does not stay neutral; someone else recruits it.

The other view — and it is strong

The best competing case is not "this is just a negotiating posture." It is that proximity itself is the leverage, and it is close to inescapable. On this reading, continental gravity — shared infrastructure, integrated supply chains, geography that makes the US the natural market for anything Canada or Mexico produces — is so overwhelming that neighbours cannot meaningfully re-route, whatever their resentment. Mexican goods still have to cross one border; Canadian energy still flows most cheaply south. Alienation changes the mood music, not the map. Concessions get extracted, relations normalise, and the advantage proves self-healing precisely because there is no realistic substitute for a next-door superpower.

What is fact, what is interpretation:

Fact: tariffs, annexation rhetoric and intervention threats have occurred; Canadian and Mexican public opinion has turned sharply against Washington.
Interpretation: whether this is reversible pressure or lasting estrangement.
Unknown: how much re-routing is actually possible given proximity, and where the trust line sits. Trust erodes without a single dramatic break; by the time it is obviously gone, the re-routing is already under way.

The deciding question is whether neighbours treat this as a storm to wait out or a reason to restructure — and whether restructuring is even feasible. Restructuring is sticky: supply chains and alliances, once moved, do not snap back when pressure eases. But the competing view's wager is that they can barely move at all.

Where else this shows up

The pattern — a durable advantage mistaken for a permanent one, then spent as leverage — recurs:

A dominant employer cuts conditions, assuming staff can't easily switch. They stay, resentfully, until a rival appears — then leave all at once, with the knowledge the firm relied on.
A platform with a captive developer base raises fees, treating the ecosystem as owned rather than courted. Developers build for a competitor; a network effect that took a decade to grow unwinds in two years.
A marriage where one partner treats the other's commitment as fixed background rather than something renewed. The commitment was real — and contingent — and quietly being withdrawn.

The common error is treating a relationship as a possession. Possessions stay put. Relationships respond to how they're treated, often with a lag that hides the damage until it's done. The counter-case is real too: sometimes switching costs genuinely trap the other party, and the advantage is near-permanent. Which world you're in is an empirical question — not an assumption to make for free.

What to watch

Two signals separate the two stories:

Trade diversification. Watch whether Canada and Mexico sign and deepen trade and security deals that cut US dependence — with the EU, Asia, each other. Concessions under pressure reverse; structural re-routing doesn't.
Cooperation on shared problems. Watch whether joint work on crime, migration and border management continues or quietly stalls. That cooperation is the subsidy in action; its withdrawal is the tax arriving.

Expectation: ~60% that within five years, trade and diplomatic data show Canada and Mexico measurably diversifying away from the US — a partial, lasting erosion of the continental base rather than a full return to the prior relationship. If instead the data show rapid normalisation and no durable re-routing, the "proximity is inescapable" view wins, and the model's warning was overstated. The geography is permanent. The advantage it confers is not.

Distilled from Foreign Affairs