didn't fire his gun." The store owner may find herself entertaining an unpleasant counterfactual because she was one day late in the installation of a new security system. Finally, a police officer, viewing the tape from the store's surveillance camera, may think, "If only the robber had looked up long enough for us to get a good view of his face." The role a person plays in a given situation affects his or her knowledge, focus of attention, and so on, all of which in turn affect the mental availability of events and possibilities and the resultant counterfactuals that are considered. To a lesser extent, a person's role may determine how natural-law constraints affect his or her counterfactual thinking. A person with sophisticated knowledge in an area of expertise (e.g., physics) probably has a very different conception of which events are reasonable to mutate than does a person with less knowledge or experience.
Examples such as these led us to reject postulating a strict hierarchy of constraints on counterfactual thinking. At first glance, it may appear that there is a natural order of constraints: Natural-law constraints are automatically triggered and restrict an infinite set of possible mutations to a relevant subset from which the most available events are most likely to be mutated. Purpose constraints further restrict the selection of counterfactual thoughts that are overtly expressed. Clearly, however, a person's purpose or role can affect the availability of events and, thus, which events are subsequently mutated. Such exceptions may call a general hierarchical structure into question. At the very least, they suggest that the constraints on counterfactual generation and selection may not operate unidirectionally.
The concept of counterfactual constraints provides a useful framework for the evaluation of past research in the area of counterfactual thinking and contributes some new ideas that are worthy of empirical investigation. There is clearly room for elaboration of the dimensions that distinguish the categories, for the inclusion of novel categories, and for the inclusion of more variables within the categories. In addition, an exploration of the distinctions between the processes and the outcomes of counterfactual thinking would be beneficial. Purpose constraints are especially interesting in this regard, because they affect both the generation of counterfactuals and the expression of counterfactual thoughts to others.