Ought Implies Can

The principle “ought implies can” asserts, roughly, that an agent ought to do something only if it is possible for that agent to do it. In other words, it's being possible for an agent to perform a certain action is a [Page 1555]necessary condition on that agent's having an obligation to perform that action. In this way, the principle “ought implies can” is fruitfully understood not as a moral principle itself, but as a principle to which ethical theories must adhere. That is, for a candidate ethical theory or moral principle to be justifiable, it must not violate the principle that ought implies can; it must not require that agents perform actions they simply cannot perform. In this way, the principle itself recognizes that ethics is concerned with action, not simply with positing theoretical ideals in accordance with which no human agent can act.

Beginning with Immanuel Kant, most moral philosophers have found some version of this principle to be intuitively obvious, even if only because its denial gives rise to apparent absurdities. For example, to deny that ought implies can is to suggest that an agent (alone) could currently have an obligation to eradicate global poverty, even though that agent has no means currently available to do so. As Kant saw it, for the moral law to have any force over human agents, it must be possible for those agents to act in accordance with the duties that arise out of the moral law.

This principle is also useful in identifying unreasonable calls for businesses to reform their practices, such as demands that automakers radically (and immediately) decrease emissions and increase fuel efficiency, even when the necessary technology is not available. It might even be the case that, were the necessary technology available, nonetheless the demand could not reasonably be met; the implementation of the technology might be prohibitively expensive, forcing the automakers to ignore their prior obligations to shareholders (e.g., to provide a reasonable return on their investment) or to consumers (e.g., to provide goods and services at a fair price).

As the preceding example illustrates, for all its intuitive appeal, there are difficulties in identifying the relevant sense of “can” in the “ought implies can” principle. There are at least two leading candidates: (1) what the currently available (to the agent) resources and other means make possible and (2) what the laws of nature allow.

The first of these seems to be the operative sense of “can” when it is claimed, for example, that an individual agent has no moral obligation to repay a debt, absent the resources for doing so. This version, however, seems to ignore prior acts and/or omissions that make the current lack of resources itself morally blameworthy and insufficient to suspend the obligation to repay. Thus, even though the debtor currently is not able to repay the debt, it might nonetheless make sense to blame—and even punish—the debtor for the failure to fulfill the obligation to repay. Doing so would acknowledge that there were actions the debtor could have performed since incurring the debt that would have made repaying the debt now possible. (To return to the previous example, it might be that automakers are partly to blame for the lack of currently available, reasonably priced technologies that can control emissions and increase fuel efficiency. Perhaps they ought to have been pursuing such technology more aggressively before now.)

...