• In the brief section following his description of these problems (pp. 48-52), Okasha notes how an account of explanation based on the idea of causality might avoid them. Although the association of explanations with the relation between causes and effects is very old, it has been most closely identified in recent decades with the work of Wesley Salmon (1925-2001). Okasha’s suggestions for further reading at the end of the book don’t suggest who in particular he may have in mind when he speaks of causal theories of explanation, but Salmon emphasized the issues of asymmetry and irrelevance (and the example of John, who takes birth-control pills, seems to have been his invention).

Okasha concludes this section by looking at one possible objection to causal theories, which suggests that they leave out certain sorts of explanation. You should think whether he is right about the sort of example he gives—and also think whether there are other sorts of examples. And ask yourself whether Hempel’s approach is more successful in handling the examples Okasha mentions.

• Another way of understanding explanations starts from the fact that they serve as answers to why-questions. The article by Bas van Fraassen (1941-) presents one version of this approach. Although he was almost a generation younger than Salmon, their ideas about these issues developed around the same time, and they reacted to each other’s views. (You will see van Fraassen referring to an earlier stage in Salmon’s thinking in which he placed emphasis on certain probabilistic relations; the birth-control example was actually devised to support this view.)

The most important parts of van Fraassen’s article for our purposes are the last two sections (§III.5 and §IV on pp. 149-150). He presents his account of explanation in the first and comments on the role of explanation in science in the second. It would be fine to start there, but be aware that you will need to read the first part of the article to understand all the ideas he uses in §III.5. Most of these ideas are outlined in sections II.4-III.4 (pp. 146-149), and it would be possible to look back at them as needed while reading §III.5.

The first part of the article (§§I-II.3, pp. 143-146) gives a survey of views of explanation that is interesting in its own right, so it makes equally good sense to just read the article straight through. But, if you do that, don’t get bogged down by earlier discussions: remember that the key points show up on the last two pages.