Basic research and why it's important

Recently, Physics Today, flagship publication of the American Physical Society (but not a refereed journal), published an interesting article on "soft," liquid or liquid-like states of matter (requires subscription). I was struck by how much progress has been made in the last 30 years on the subject and by how much that progress was due to the interaction of applied and basic, or pure, research. Such achievements show better than any abstract argument why science can't function without both.

Basic research is not necessarily theoretical -- it can include experiment -- but it is conducted with no immediate application in mind. It's obvious enough that pure research needs both experimental and applied research. (Ignore this, and you get string theory.) I don't think any successful pure research was ever not a result, direct or indirect, of some attempt to solve an applied problem.*

But applied research needs pure research as well. The reason is that applied research, carried far enough, encounters or poses problems that applied research alone cannot answer. Re-posed as pure research, these problems and their eventual solutions in turn often have unexpected and broad ramifications. There's Fourier analysis, for example; 200 years ago, it was mainly a mathematician's curiosity. But by the late 19th century, it was already very practical. Today, most of the technological gadgets we encounter day to day owe something to Fourier's discovery.

Weather and climate constitute a case where understanding beyond a certain point fails because of large, unanswered basic questions about the definition, nature, and prediction of "climate" over long times. The pressure on scientists to resolve questions that currently lack answers is what has produced the groundless para-science of "global warming." Without the real thing, empty mummery and BS fill the void. A "consensus" is manufactured. Pressure to conform to this consensus overrides scientific standards and critical thinking. Demands for policy-ready conclusions are a major culprit.

POSTSCRIPT: With hurricane season upon us, making climate predictions more useful is becoming topical. But typical weather predictions of a few days or the two-week type are all that's needed to get people out of the way of incoming storms.** There's no point in creating an industry of bogus "long-term forecasting." At the present stage of the science, such an industry can be no better than reading tea leaves, beyond obvious seasonal and other astronomical patterns.

The value of basic research arises from the apparent paradox that, to be useful in the long run, it has to be "useless" in the short.
---
* That's true even of subatomic physics, which has its origins in 19th century theories of electromagnetism and early 20th century attempts to combine electromagnetism and quantum mechanics.

** Even better is finding ways of discouraging people from living in likely storm paths in the first place -- or at least, not encouraging them.

Wealthier countries have early warning systems in case of severe weather. A helpful step would be to make these early warnings more available in poorer countries. But such steps have nothing to do with improving weather forecasting or climate science. They're about making already-known information better disseminated and, in that sense, "more useful."