For much of the twentieth century, economists treated the depletion of shared resources - fisheries, grazing lands, irrigation systems - as an inevitability rooted in individual rationality. The canonical formulation held that when a resource is jointly accessible and finite, each user, seeking to maximize private benefit, will extract as much as possible; because the costs of overuse are distributed across all users while the gains accrue privately, restraint is irrational for any single actor. The predicted outcome - ruin shared by all - seemed to license only two remedies: privatization, which assigns exclusive ownership, or centralized state control, which imposes extraction limits from above. Empirical work beginning in the 1970s complicated this dichotomy. Researchers documented numerous communities - Swiss alpine pastures, Japanese village forests, Spanish irrigation networks - that had governed shared resources sustainably for centuries without either private title or external enforcement. These arrangements were neither anarchic nor imposed; they were self-organized systems of rules, crafted and revised by the users themselves. The mere existence of such cases did not refute the original logic, its defenders noted, for the theory described a tendency, not a universal law, and successful commons might simply be exceptions preserved by unusual circumstance. The more consequential challenge was analytic rather than anecdotal. Studying the durable cases comparatively, scholars identified recurring institutional features: clearly defined boundaries delimiting who may withdraw; rules calibrated to local conditions; mechanisms allowing most users to participate in modifying those rules; monitoring by accountable parties; and graduated sanctions that escalate with the severity and repetition of violations. Where these features were present, cooperation tended to persist; where absent, it tended to collapse. This did not establish that the features caused durability - communities capable of devising good rules might also possess other advantages - but it reframed the question. The relevant variable was not whether a resource was owned but whether its users could communicate, monitor one another, and adjust their agreements over time. The reframing carries a caution often overlooked by enthusiasts. To show that centralized or private solutions are not always necessary is not to show that community governance is generally superior, or even generally available. The documented successes shared conditions - relatively stable populations, low-cost monitoring, shared expectations - that many contemporary resource dilemmas conspicuously lack. A global fishery, whose harvesters are numerous, anonymous, and mobile, resembles the durable village commons in almost no respect. What the research licenses, then, is not optimism about self-governance but skepticism toward any claim that a single institutional form fits all cases. The choice among ownership regimes is properly an empirical matter, contingent on the attributes of the resource and its users, rather than a deduction from first principles about human motivation. Yet the deductive habit persists, partly because it flatters the disciplines that trade in it. A model that predicts ruin from a few axioms is more portable, and more teachable, than a catalogue of contingencies. The enduring appeal of the tragedy narrative may thus reveal less about resources than about the incentives of those who study them.