Abstract: Defenders of the ontic view of scientific explanation argue that explanations for scientific phenomena are in the world regardless of whether we discover them e.g. what explains the hole in the ground is the actual meteor that hit the Earth, not our description of the meteor hitting the Earth. However, critics of the ontic view argue that it fails to capture the importance of idealized models as a critical component of scientific practice. Specifically, Robert Batterman argues that highly idealized minimal models of physical phenomenon are counter-examples to the ontic view insofar as minimal models purport to leverage explanatory power by leaving out all the ontic details. In this paper, I argue that existence of minimal modeling as a scientific practice is consistent with the ontic view of explanation.
This paper topic stems from a recent Philosophy of Science workshop I attended. At the workshop, many talks centered around a contrast between two views of scientific explanation: an “ontic view” and a “non-ontic” view (for lack of better terms). On the ontic view, scientific explanations are out there in the world and models are idealized descriptions of these explanations. On the ontic view, the extent to which a model or description explains a phenomena is proportional to the extent to which that description makes reference to the underlying ontic explanation. In contrast, the non-ontic view states that idealizations are not just “incomplete” or “partial” explanations to be filled in with more ontic details later. Rather, the explanatory work is being done by the details left out of the model.
The paper is an attempt to analyze these views and determine whether they are inconsistent. I argue that the existence of idealizations in scientific practice does not undermine the ontic view of explanation, despite Batterman and others claims to the contrary. At best, they are orthogonal. Batterman’s point is that idealizations are better explanations than ones referring to micro-levels because they help us understand the aspects of the phenomena that we are most interested in: the macro-level regularities. However, the very fact that Batterman is focused on what’s interesting is irrelevant to the ontic view because they are ontic explanations of both interesting and non-interesting phenomena. Thus, the point of what’s interesting to humans is a non-starter as an objection to the ontic view.
Rather than diagnosing this as a mere terminological dispute over the correct usage of the English word “explanation”, I appeal to Richard Feynman’s remarks on mathematics and physics to highlight an under-appreciated feature of theoretical physics that makes it distinct from mathematical physics: the notion that conjectures in physics must have a “physical meaning” to be true whereas mathematical conjectures do not.