> What we've done is essentially say that "research isn't valuable unless the researcher gets federal funds to do so, or is paid for by a private corporation."

> What we've done is essentially say that "research isn't valuable unless the researcher gets federal funds to do so, or is paid for by a private corporation." This has happened in federal republics: in the past states were more powerful and expected to bear a larger portion of the Public Good-producing apparatus. Then some stuff happens (maybe production is de-localized even if it remains intra-national; maybe states enter races-to-the-bottom in state taxes to attract industries, etc.) and only the federal government is able to deal with problems. So that's probably orthogonal to research. The basic problem: research broadly framed is a public good, like a lighthouse. But hey, math research is thoroughly public, theorems are all utterly equivalent to a small number of rules of metamathematics. OTOH drug research can often be access-controlled so it can be made profitable. This is the pure-applied spectrum, and all the innovative solutions are about having private actors pay for stuff at the margins in the applied end. For example: I work with symplectic numerical integrators, maybe Mathworks out of its enlightened self-interest should pay me so that even if the research comes out in papers it will be first on its MATLAB suite of engineering stuff. See?