The main philosophical questions regarding harm are, first, what exactly it is, and, second, whether it matters morally. The questions are connected: whether harm has moral significance must surely depend, at least in part, upon how it is understood. Not every moral theory assigns fundamental significance to harm. According to act utilitarianism, the right action to perform is the one that would result in the greatest sum of well‐being among sentient creatures. The theory attributes no independent significance to how actions affect particular individuals. Furthermore, even if a utilitarian were to take an interest in a particular individual's fate, he would ask only which of the available actions would be best for that individual. He would not care whether any of the actions harmed the individual. If A gives B a small benefit when he could instead have given him a larger one, he may not harm B, but he does not do what's best for him either; and, if A chooses the least harmful option when he cannot help but harm B, he harms B, yet he does do what's best for him. There is no necessary connection between doing what's best for someone and avoiding harming him. As a first approximation, let us say that A harms B if and only if one of A's acts or omissions negatively affects B's interests or well‐being. On this maximally broad understanding, an action's harming someone is much the same as its being bad for him. If a moral theory's interest in how an action affects someone is exhausted by the question whether its performance is good or bad for him, that theory has no special need for the concept harm. It need only employ the concept bad for. It is possible, however, to understand harm more narrowly than this, and to see the narrower notion as having moral significance. Our more public actions almost invariably affect other people's interests, if only indirectly; and, while we should perhaps always bear these effects in mind when deciding what to do, many think that we needn't in general give them all that much weight in our deliberations. But harming someone – say, by physically assaulting him – can seem a much more serious matter, morally speaking, than merely negatively affecting his interests. A deontological theory might well hold that we have an especially strong reason – perhaps even an obligation – not to harm people. Likewise, it might hold that we have an especially strong reason – perhaps even an obligation – to prevent people from suffering harm and to come to the aid of those who are already suffering it. On such a view, to harm someone is to inflict an especially important kind of negative effect upon him.
