There were times in which philosophy was considered as the highest form of education, a sort of "meta-knowledge" you acquire that enables you to reason about any other corpus of knowledge. In France it is still considered as such, although what French call "philosophy" is a wierd mixture of erudition, rhetoric capacities and "esprit" in conversation. I think that this still holds, and that studying philosophy enables you to acquire a skillful mind: not only a faster one (as in the case of studying very formal disciplines, like mathematics), but also a more reflective one.

And I do not see the "impracticality" associated with it, apart from the fact that it makes it harder to get a job (I remember a cartoon in my department in which you could see the scene of an interview of a candidate for a job and the cadidate saying: "It is true that I have a doctorate in philosophy, but I'm willing to learn"). There's something "practical" in being mindful, and the landscape of knowledge changes so fast that it is not bad to stick to good old philosophical questions, whose survival is assured in any future possible cultural or scientific transformed scenario.

I wonder what is meant in the question by talking of philosophy as a self-perpetuating field? In what sense is philosophy supposed to be "self-perpetuating" while biology isn't?

Perhaps the idea is supposed to be that philosophy is self-perpetuating because, unlike biology, it just goes round in circles for ever and never settles anything. If that is the implicit claim, then I think it should be resisted vigorously. It would be just absurd to deny that we now know a vast deal more about issues about language, meaning and reference than we did before the time of (say) Frege; it would absurd to deny that we now know a vast deal more about the nature of the mind than we did before the time of (say) William James. Again, think about the philosophy of space and time: it would crazy to suggest that we are stuck where Newton was. And so it goes, through area after area. Of course, "settling a question" in philosophy isn't exactly like settling a question in biology (though that too, as the philosophers of science remind us, can be a more complex business than we might suspect). Often, what becomes clear are the costs and benefits of accepting this or rejecting that, sometimes questions dissolve on careful enquiry, or other questions shift as our related scientific knowledge advances, and so on. But the fact that progress in serious analytical philosophy is a complex business certainly doesn't mean it doesn't happen.

And I'd perhaps rather resist too Gloria Origgi's talk about the "good old philosophical questions", for that way of talking too readily suggests that the problems don't shift very significantly over time.

Still, even if philosophy isn't "self-perpetuating" in the bad sense of just going fruitlessly round in circles, you might say that it is still "impractical." Well, thinking about the kind of foundational issues in the sciences that continue to feed into the most lively areas of philosophy (the philosophy of biology for one!) may not give immediate "practical" pay-offs. But it would again be absurd to deny that foundational enquiries have often fed back in the end into the growth of knowledge of the most practical kind. Just think, for example, how Turing's work thinking about the very notion of a computation led to modern computer science.