Luckily I got into graduate school for a great program in Europe where people tend to be more well rounded, but my advisers were right in their predictions in that I never got into a graduate program in the USA (whereas in Europe several were interested, although in some interviews the public outreach WAS addressed as a liability even here). It just worried me that a field that de facto relies on taxpayer money would consider someone interested in explaining science as such a bad thing.

Without evaluation of the actual quality of research, there is little reason to gripe about breadth or depth. A crummy paper doesn’t become good because it is interdisciplinary.
Writing five bad papers on five separate topics does not make you better than someone writing five good papers on one topic.

It is true that brilliant, radical and foundational insights may come from another discipline. And it is also true that the truly great people (thinking of people like Maldacena, Weinberg, Wilczek, Witten, etc) leave contributions in many disciplines. But the converse isn’t true. This is the Galileo fallacy – Galileo’s ideas were rejected, my ideas are rejected, therefore I am Galileo – and writing papers across different areas does not a great thinker make.

At least in physics, work that sets out to be broad (‘Today I am going to write an inter(sub)disciplinary paper’) on the whole does not seem to me that good. The best work of this kind is where the link has emerged by naturally following good ideas.

Times have changed. A few years ago, someone was awarded a doctorate for a thesis involving Pirsig’s work.

It seems that increasingly departments and individual researchers are playing fetch with whatever funding agencies decide to throw into the field. One way to halt the trend of ever increasing specialization is to de-compartmentalize funding agencies such as the NSF.

But on a more objective level:
I understand the argument for specialisation/depth — that with our current state of knowledge, it is very hard to make significant progress any other way — but does that mean *every* academic needs to be ultra-specialised? Many academics spend an entire career without doing any ground-breaking work anyway, so perhaps it would be beneficial if some positions were instead taken up by broad thinkers, who can look at the bigger picture, and assist with cross-fertilisation of ideas etc., maybe without ever making any earth-shattering discoveries themselves. Perhaps I am being idealistic, and this doesn’t really work in practice…

The problem runs deeper than that. That academia passively and actively discourages interdisciplinarity has the effect that scientists get stuck in a discipline they once thought promising. Overspecialized as they are, the only choice they have is to continue to proclaim their research field is still promising.

Specialization occurs naturally, in research as well as in ecological systems. It’s a division of labor that allows to use resources best. But it’s not useful to discover new resources. This often comes from making connections between different fields. There’s an interesting paper by Alexander Shneider who argues that science proceeds in 4 stages and specialization is one of them. That’s of course a gross oversimplification of the complexity of knowledge discovery, but I think there’s some truth in it.