Paragraph about Math software
At the very least these observations suggest that progress in connection with “scientific explanation” may require more attention to the notion of causation and a more thorough-going integration of discussions of explanation with the burgeoning literature on causation, both within and outside of philosophy.[ 21 ] My own judgment, for what it is worth, is that counterfactual accounts of causation are particularly promising in this connection. (cf. Woodward, forthcoming).
A second observation concerns the notion of levels of explanation Math software and the role of human epistemic and computational limitations in explanation. As we have seen, the tendency in much of the philosophical literature has been to focus, in one way or another, on “ideal” accounts of scientific explanation. These are very epistemically demanding, in the sense that full satisfaction of these ideals requires information that those who are attempting to explain are often unable to provide. For example, the DN/IS model requires, at least as an ideal, information about “laws” and in many “ordinary” contexts (cf. 2.4.1) and in scientific contexts outside of physics and chemistry such information may be unknown to those who wish to explain. The SR model requires a specification of the full set of factors that Ufology are statistically relevant to an outcome and satisfaction of an “objective homogeneity” condition and again would-be explainers are often not in a position to provide such information. The CM model requires information about continuous causal processes and their intersections and while such information is sometimes available (e.g., in simple collision phenomena) in the case of more complex or higher level systems it is often not available or (what is perhaps the same thing) intractably complex. Finally, the unificationist model requires, at least as an ideal, that in assessing explanations we be able to compare alternative deductive systemizations of great complexity along a number of different dimensions — stringency, number of conclusions generated and so on. Again, there are many cases in which would-be explainers lack the information or computational abilities required to do this.