It's not that surprising when you consider that the English-speaking societies have been getting increasingly ideologized for over a century. By this I mean both the increase in ideological uniformity and the tendency to see an increasingly narrow range of topics as non-ideological and thus freely debatable. From the official intellectual institutions, we still get interesting ideas about strictly technical topics that don't involve any ideological issues, but otherwise, these institutions are nowadays hostile to anything but conformist intellectual cant. So it's no wonder that insofar as some real intellectual life persists, it subsists only in obscure underground venues.
(The situation is in fact similar to the former Soviet Union, where the hard sciences worked reasonably well, but one would be foolish to look for much beyond official ideological cant from, say, historians or sociologists, and real intellectual life was happening underground. Of course, the ideological uniformity in the U.S.S.R. was enforced in much cruder and harsher ways, but the more subtle carrots and sticks that the present system offers to intellectuals are, if anything, only more effective.)
Actually, by some metrics the USSR's hard sciences did pretty abysmally also. See Hyena's remarks in this subthread here.
SIAI benefactor and VC Peter Thiel has an excellent article at National Review about the stagnating progress of science and technology, which he attributes to poorly-grounded political opposition, widespread scientific illiteracy, and overspecialized, insular scientific fields. He warns that this stagnation will undermine the growth that past policies have relied on.
Noteworthy excerpts (bold added by me):
In relation to concerns expressed here about evaluating scientific field soundness:
Grave indictors:
HT: MarginalRevolution