Abstract: Something akin to what social psychologists call the “fundamental attribution error” underlies many discussions of the responsibility of politicians – from Donald Trump to Jacinda Ardern – or the bad or good outcomes the nations they led had with the COVID-19 pandemic. Observers saw what the leaders did, and saw the outcomes. The fundamental attribution error is a tendency to explain the outcome more by pointing to what the leader did than to the context in which she acted. This Essay argues that we have to understand social events as the interaction between human agency and the constraints under which people act. The widespread governance failures in responding to the coronavirus pandemic tend to generate accounts that overemphasize agency and underemphasize constraint. The very scope of the failures – that only a handful of governance mechanisms around the world generated policies that did a decent job of keeping COVID-19 under control – suggests that we should look more closely at the constraints under which policy-makers operated.This Essay uses the distinction between agency and constraint as a tool for helping us think about the policy responses that were available and likely to be used in early 2020, when the “novel” coronavirus came on the international scene. The bottom line is this: given the context within which policy-makers acted (the constraints they faced) as the crisis developed, the pandemic was quite likely to be a human catastrophe. It’s not that nothing could be done to stop it, or even that nothing could be done to make it “merely” a disaster instead of a catastrophe. And it’s not that no one came up with – and sometimes implemented – policies that helped limit the disaster’s scope. The constraints under which policy-makers operated, though, meant that the chances of really successful outcomes were quite low – a suggestion consistent with the fact that outcomes around the world were basically pretty bad.The Essay proceeds by first identifying major features of the context as of early 2020 – the constraints and context for policy-making. Part II then describes what we know now, or have strong reason to believe, were the policies that could have done the most to minimize the virus’s effects on life, health, and economies. Part III examines the choices that were actually made, focusing, for reasons to be discussed, on nations with generally democratic systems of governance. A brief Conclusion returns to the fundamental attribution error: context and constraint probably mattered more than agency in generating the bad outcomes around the world.