THE TAIWAN CALCULATION: Part I
Two Ways to Break the Rules
Baghdad, March 20, 2003. 05:34 local time.
The first cruise missiles hit the Iraqi capital before the ground invasion began. In the hours that followed, American and British forces crossed the Kuwaiti border in three columns. The operation had a name: Shock and Awe. The idea was simple. Move fast enough, and the enemy’s will to resist collapses before it can organize. Saddam Hussein’s government fell in twenty-one days.
The United Nations Security Council had not authorized the invasion. France and Germany had made clear they would veto any resolution that did. Washington went ahead anyway. The legal justification, weapons of mass destruction that were never found, was secondary to the strategic logic: a hostile regime in a pivotal region, a perceived window of military superiority, and a calculation that the cost of waiting exceeded the cost of acting.
The world condemned it. And then, largely, moved on.
Eleven years later, on a cold February morning in 2014, Russian special forces without insignia occupied government buildings in Simferopol, the capital of Crimea. There was no declaration of war. No Security Council resolution. No shock and awe. Within days, a referendum was organized. Within weeks, Crimea was annexed. The operation took less time than it takes most governments to schedule a cabinet meeting.
The world condemned it. And then, largely, moved on.
Two operations. Two great powers. Two violations of the international order that the postwar world had built over seventy years. And in both cases, the consequences for the aggressor were manageable. Sanctions, diplomatic isolation, historical opprobrium. But no reversal. No meaningful punishment. The facts on the ground remained facts.
This is the first thing Beijing noted.
Not the moral dimension. Not the legal arguments. Not the editorials in the international press. The operational conclusion: when a great power moves decisively, and moves fast enough, the international order absorbs the blow and reconfigures around the new reality. The rules do not enforce themselves. They are enforced by states with the will and the capacity to enforce them. And that will, as both Washington and Moscow had demonstrated, is conditional.
But Beijing noted something else as well. Something more uncomfortable.
Iraq did not end in 2003. It ended, in a certain sense, in the years of occupation that followed, in the insurgency that consumed American resources and credibility, in the sectarian war that Washington had not planned for and could not control. The military victory was total. The political outcome was a generation of strategic damage.
Ukraine, similarly, did not end in the first weeks of February 2022. Russia expected it to. The same logic that had worked in Crimea, speed, ambiguity, fait accompli, was applied to a full invasion. It did not work. Eight hundred days later, Russian forces held parts of four Ukrainian oblasts and had lost more men than in any conflict since the Second World War. The military operation that was supposed to last weeks had become a grinding war of attrition that exposed every structural weakness in the Russian state.
Two models, then. And two very different outcomes.
The first model: limited, fast, defined objective, fait accompli before the international system can respond. Crimea 2014. And, as we will see, Iran 2025. Costly in diplomatic terms. Manageable in strategic terms.
The second model: full invasion, regime change ambition, assumption of rapid collapse. Iraq 2003. Ukraine 2022. Militarily successful in the short term. Strategically catastrophic in the medium term.
Beijing has studied both. Not as history. As a planning document.
The question is not whether China will act on Taiwan. The question is which model it will choose. And whether it has found a way to combine the speed of the first with the ambition of the second, without inheriting the failures of either.
That question will not be answered in the conference rooms of international organizations. It will not be answered by diplomatic communiqués or carefully worded Security Council resolutions.
It will be answered, if it is answered, somewhere over the Taiwan Strait. Probably on a morning that, in retrospect, will not have come as a surprise to anyone who had been paying attention.
This series is an attempt to pay attention.
The Anatomy of Model One
To understand what Beijing is watching, it helps to be precise about what made the first model work, and what made it work only under specific conditions.
Crimea succeeded because the objective was narrow and the timeline was short. Russia did not try to change the Ukrainian government. It did not attempt to occupy Kyiv. It took one peninsula, of specific strategic value, the home of the Black Sea Fleet, the warm water access that Russian grand strategy had coveted for three centuries. The operation was designed to be irreversible before the international community finished drafting its response.
Speed was not incidental. Speed was the strategy.
And it worked because the target could not resist effectively, because NATO had no mandate to intervene, because the United States under Obama calculated that the cost of military confrontation over Crimea exceeded the cost of absorption, and because Europe, dependent on Russian gas, was structurally inhibited from escalating. Every one of those conditions was specific to that moment, that geography, that constellation of interests.
The lesson was not: limited operations always work. The lesson was: limited operations work when the target is isolated, the objective is defined, and the defender’s alliance cannot coordinate a response faster than the operation concludes.
Iran 2025 follows the same logic, updated for a different theater. The strikes on Iranian nuclear facilities did not aim to change the regime in Tehran. They aimed to degrade a specific capability, the path to a deliverable nuclear weapon, by a specific number of years. The objective was surgical. The timeline was measured in hours, not months. And the international response, however loud, arrived after the operational facts were established.
Washington, like Moscow in 2014, calculated that the cost of waiting, a nuclear-armed Iran, exceeded the cost of acting, diplomatic isolation and regional instability it could manage. The legal basis was contested. The strategic logic was clear.
Two powers. Two different decades. Two operations that share one underlying structure: act before the system can stop you, limit your objective to what you can hold, and absorb the diplomatic cost from a position of accomplished fact.
The Anatomy of Model Two
The second model shares the same starting logic. The cost of waiting exceeds the cost of acting. But it adds an ambition that the first model deliberately excludes: the transformation of the target.
Iraq was not about removing a weapons program. It was about removing a regime and replacing it with something compatible with American interests in the Middle East. That ambition required not just military victory, which came quickly, but political consolidation, which never came at all. The assumption was that Iraqi society, liberated from Saddam, would organize itself around something stable and pro-Western. That assumption was wrong in ways that the pre-war intelligence, and the pre-war politics, did not allow to be examined seriously.
Russia made an almost identical error in February 2022, with almost identical consequences. Moscow assumed that Ukraine’s political identity was shallow, that the Zelensky government would collapse or flee within days, that Russian-speaking Ukrainians would welcome, or at least not resist, the return of Russian influence. Every one of those assumptions was wrong. Ukraine fought. The government stayed. The population, including large parts of the Russian-speaking east, chose resistance over accommodation.
The military planners had prepared for a campaign. They found themselves in a war.
The structural failure in both cases was the same: the aggressor confused military superiority with political outcome. Destroying an army is not the same as destroying a state. Occupying a capital is not the same as pacifying a country. The second model works only if the target society is willing, or too broken, to resist the new order. Neither Iraq nor Ukraine met that condition.
What the Record Actually Shows
Set the moral arguments aside for a moment. Not because they are unimportant, but because Beijing does not make strategic decisions on moral grounds, and neither, in practice, does Washington or Moscow. What the record of these four operations actually shows is the following.
Great power unilateralism is survivable. None of these operations ended the international standing of the state that conducted it. The United States remains the world’s leading power two decades after Iraq. Russia, diminished and sanctioned, still holds Crimea. The rules-based order bent. It did not break.
But the second model carries an existential risk that the first does not. Full invasion with regime change ambition can trap a great power in a conflict it cannot win and cannot exit. It can consume resources, credibility, and strategic attention for a generation. It can expose every internal weakness to external observation.
The first model, by contrast, is reversible only in theory. In practice, the international community has demonstrated, twice now, that it will not reverse an accomplished territorial or strategic fact through force. It will sanction, condemn, and contain. It will not undo.
For a strategist in Beijing, reading this record without sentiment, the conclusion is uncomfortable in its clarity. The first model works. The second model is a gamble. And the question that will shape the next decade of international order is whether Taiwan represents, in Chinese strategic thinking, a Crimea or an Iraq.
The answer to that question is not yet written. But the inputs into Beijing’s calculation are visible, if you know where to look.
In the next part of this series, we will look at exactly that: what China’s military and political leadership has drawn from these four cases, and why the synthesis they have developed is more sophisticated, and more dangerous, than either model that produced it.
The Taiwan Calculation continues with Part II: What Beijing Learned.