Research
Some true explanations are bad explanations. A true explanation can be too specific, too abstract, too unilluminating, or simply pitched at the wrong level. Consider a pigeon that has been conditioned to peck when presented with red stimuli and, separately, conditioned to peck when presented with food. On a particular occasion, it is presented with a scarlet stimulus and pecks. Why did it peck? We might say: because it was presented with a scarlet stimulus. We might say: because it was presented with a red stimulus. Or we might say: because it was presented with either a red stimulus or food.
All three answers may be true. And each has something going for it. The first answer is the most precise: the stimulus was not merely red, but scarlet. The third answer captures a broader generalization: red stimuli and food are both routes to pecking. Yet the second answer seems best. It says enough, but not too much. It abstracts away from the irrelevant detail that the stimulus was scarlet, but it does not generalize so far that it lumps color-based pecking together with food-based pecking.
But why? What makes one true explanation apt, while nearby true explanations are too fine-grained or too coarse-grained? This is the Goldilocks problem for explanation. I argue that the problem is not merely a local puzzle about causal explanation or the special sciences. It is a general problem about explanatory aptness: about why some ways of describing a phenomenon are fit to explain it, while others are not.
Explanatory Aptness and Inquiry
In preparation
Why can two true answers to the same why-question differ in explanatory quality? I argue that an explanation is apt when it resolves the live contrast organizing an inquiry at a particular stage. Inquiry is structured by tasks, background information, and inferential bottlenecks; an apt explanation addresses the distinction that must be settled for inquiry to proceed. The view captures the context-sensitivity of explanation while retaining objective standards of relevance.
The Explanatory Goldilocks Problem
In preparation
Good explanations often seem to occupy a middle position. They leave out irrelevant detail without abstracting so far that they group together importantly different possibilities. This paper formulates that problem in general terms and argues that familiar appeals to counterfactual range, precision, or unification do not by themselves explain why one level of description is “just right.” The problem is not limited to causal explanation or any single scientific domain.
Explanatory Aptness and the Autonomy of the Special Sciences
In preparation
Debates about explanatory level are often linked to the autonomy of biology, psychology, economics, and other special sciences. I argue that this connection is less straightforward than it first appears. In ordinary Goldilocks cases, the apt level may shift with the inquiry. By contrast, higher-level scientific explanations can earn their explanatory value by identifying robust patterns that remain stable across lower-level variation. Distinguishing these sources of explanatory aptness clarifies what a defense of higher-level explanation must establish.