In preparation for an oral history interview I am going to be doing tomorrow, I have been reading a 2004 interview a couple of other historians did with John Armstrong, a major figure in IBM research, who had some considered and very interesting opinions on a “Golden Age of Industrial Research” in the postwar era. By his reckoning, idealism about this age is highly misleading. What he believes has generally driven the idea that it was “golden” is that academic scientists were able to identify industrial research as being something that resembled their own work and that absorbed their own (increasingly large) supply of students (Dave Kaiser–once again–has some good upcoming work on the intellectual ramifications of the postwar physics “bubble”). It is these academics who have typically spoken and written the most, and thus set the terms of discourse.
He claims, though, that this recognizably academic-style arrangement within an industrial setting did little for American industry and economy (that could not have had a more substantial impact on the nation in an alternative setting), because researchers given free rein were so frequently out-of-tune with company goals. It is his claim that this point-of-view is not controversial within industrial research and management culture, but that academic scientists still hold it to be true (update: on practical relevance of this view for current education, it is worth looking at his piece, “Rethinking the Ph.D.” in the summer ’94 issue of Issues in Science and Technology).
Armstrong attributes the postwar arrangement to a misplaced faith in the linear model of science and technology, where investment in pure research is supposed to lead directly to technological applications. This argument, however, feeds what could very well be a common misperception–David Edgerton has argued that the “linear model did not exist”. What Edgerton means by this is that the idea that postwar institutional arrangements and science policy were based on an idea that science flowed downhill into technology is false. No important figure actually held such beliefs. A more typical point of view was that it was a good idea for applied science communities to have access to more theoretical forms of knowledge, and vice versa. Yet, the institutionalization of these arrangements–whether industrial laboratories should have their own “basic research” staff; or closer liaison should be kept with academic laboratories–remained undecided. Armstrong clearly prefers more recent arrangements favoring industrial concentration on development, with more basic research based in the academy. (I defer discussion of “what is basic research, and is it OK to talk about it in such terms” to another time.)
Of course, this issue is not cut and dry since industrial secrecy requirements could hamper the value of basic research for industrial laboratories, but I’ll reserve discussion of these issues for another time, since there are some interesting trans-Atlantic comparisons that deserve exploration in depth.