First he cites the bromide, science asks “how” questions, religion asks “why” ones.
It sounds like a clear enough distinction, but maintaining it proves to be very difficult indeed. Many “why” questions are really “how” questions in disguise. For instance, if you ask: “Why does water boil at 100C?” what you are really asking is: “What are the processes that explain it has this boiling point?” – which is a question of how.
Critically, however, scientific “why” questions do not imply any agency – deliberate action – and hence no intention. We can ask why the dinosaurs died out, why smoking causes cancer and so on without implying any intentions. In the theistic context, however, “why” is usually what I call “agency-why”: it’s an explanation involving causation with intention.
So not only do the hows and whys get mixed up, religion can end up smuggling in a non-scientific agency-why where it doesn’t belong.
This means that if someone asks why things are as they are, what their meaning and purpose is, and puts God in the answer, they are almost inevitably going to make an at least implicit claim about the how: God has set things up in some way, or intervened in some way, to make sure that purpose is achieved or meaning realised. The neat division between scientific “how” and religious “why” questions therefore turns out to be unsustainable.
That’s very useful.
It’s funny, too, that people do that. The idea is that a mega-meta-person is a more satisfying answer than a mere process or brute fact. But why is it? Given that you can ask “why” about the mega-meta-person, you would think that answer would be satisfying only for a few seconds, or a few minutes for the indolent. I don’t really get why “it just did” or “no one knows yet, but people are looking” is less satisfying than “a mega-meta-person did it.” Not to mention that the latter is a great deal less plausible than the former.