A couple thoughts I had on this episode:

1) The description of the fairies creating and destroying the world as we walk through it reminded me a bit of “The Langoliers”, probably my favorite short story by Stephen King, and a unique take on time travel.

2) Regarding the idea of modern science being both empiricist and rationalist, I agree with the broad strokes of your description. Often, it’s true, theories are posited, and only later are confirmed or refuted through experiment. However, I think most scientists would agree that if a theory doesn’t conform with observations, that it should be abandoned or at least modified. Of course, this doesn’t stop many scientists from holding on to theories long past their sell-by date, and in this regard I do agree with Kuhn. Scientists are only human, of course, and nobody likes to see something they’ve worked on for their whole life be trashed, but I maintain that a good scientist will accept such a situation with grace. I disagree with Kuhn a bit when he says that it takes the old generation of scientists to die off before a new scientific paradigm is accepted. Maybe it does in many cases, but certainly not all. Einstein’s theory of relativity was, as I recall, accepted rather quickly (relatively speaking 😉 ) by physicists once the observations started coming in that confirmed several of its key predictions.

On the flip side, however, it’s also possible for an empirical observation to be made for which there is no current viable theory. A contemporary example of this is the apparent faster-than-light neutrino discovery. No real theory predicts such a phenomenon, and assuming that it is real (which it probably isn’t), will require radical rethinking of current theories. Situations like this are relatively rare, however, and there’s an excellent chance that the measurement in this particular case is erroneous (some new information that just came in today suggests that it might be an instrument error after all), but the jury is still out.

Finally, there are some scientists who argue that some theories out there today that are usually considered scientific (in the modern sense) are actually not valid scientific theories at all, primarily because they make no empirically testable predictions. The most obvious contemporary example of this is String Theory, which is an elegant mathematical framework that *could* explain all of quantum mechanics, particle physics, and so on, but the energy levels required to actually test any of its predictions are so far beyond anything we have at our disposal that it essentially can’t be tested any time in the foreseeable future. Should String Theory (or M-Theory as it is currently being called) be considered a valid scientific theory, even though it seems like it can’t be tested? I don’t know. I will say that I’m inclined to lean toward theories being considered scientific (again, in the modern sense) if they stand a reasonable chance of being tested by empirical observations, and I think that many scientists would agree with me (except, of course, the String Theorists!). Thus, I think that modern science leans toward empiricism in the sense that at the end of the day, experiment and observation trump theoretical predictions, but in the interim, there is always an ongoing interplay between the two that keeps things interesting!