Given some changes in the world, the frame problem is the problem of determining that most things in the world haven't changed. Since change is generally taken to mean "change over time", the frame problem is generally assumed to be a problem of temporal reasoning, and most examples of the frame problem are couched in terms of the effects of actions. In this paper, I point out the fallacy underlying this approach, and demonstrate something very much like the frame problem that is completely independent of time: the counterfactual validity problem. I show that this "atemporal frame problem" proves damning for certain approaches to solving the frame problem. However, exploration of the differences between these two problems may help to identify more promising research directions.