What do universities, journals, and governments need to do to stimulate breakthrough scientific discovery? originally appeared on Quora – the knowledge sharing network where compelling questions are answered by people with unique insights.
Scientific progress cannot always be described in terms of breakthroughs, and sometimes a finding is not realized to be a breakthrough until decades after the fact. But the best thing government, universities, and journals can do is to give adequate support to scientific research and not get in the way too much.
This morning, I went on a run around the Charles River and kept track of my pace and distance using a GPS sports watch, a product which one could first purchase 10-15 years ago. GPS technology relies on very accurate timekeeping, and GPS satellites each have several atomic clocks (using Cs-133 and Rb-87) which set a precise frequency standard (frequency is inverse of time) from electronic transitions between atomic energy levels split by hyperfine interactions. Hyperfine splitting was first observed by Albert A. Michelson (who won the Nobel prize for disproving the presence of aether) in the 1890s, but it was not explained until 1924 when Wolfgang Pauli proposed the existence of nuclear magnetism (there is also a contribution from the nuclear electric quadrupole moment). With nuclear magnetism as a starting point, hyperfine coupling is pretty straightforward to derive using the tools of electromagnetic theory and quantum mechanics developed in the 19th and early 20th centuries, respectively. The idea of using nuclear magnetism as the basis of an atomic clock originates from I. I. Rabi , who was awarded the Nobel prize in 1944 for his work in measuring nuclear magnetism. In addition to being used for precision measurements, the phenomenon of hyperfine coupling gives rise to one of the most useful beacons in radio astronomy—the 21-cm hydrogen line—demonstrating that sometimes the application of curiosity-driven science is more curiosity-driven science. Finally, GPS technology is perhaps the most celebrated application of Einstein’s theory of general relativity (1915), as the clocks on the satellite have to be corrected relative to observers on Earth.
The story above serves to illustrate important facts about scientific breakthroughs. First of all, they are often serendipitous. Michelson had no idea about nuclear magnetism or hyperfine interaction. His optical observations of hyperfine splitting were not the finding he became best known for, and he certainly did not seek out the seeds of consumer technology which would one day allow hapless vehicular tourists to navigate an unknown city. Secondly, they are sometimes esoteric and their application to technology and/or basic science is unpredictable. I don’t really think that the University of Hamburg (Pauli’s institution in 1924) put out a breathless press release promising quantum mechanical clocks originating from their professor’s recent paper. Nor do I see a bureaucrat in Washington DC or an industrialist in Detroit in the 1930s deciding that nuclear magnetism is an important strategic investment (it’s also the operating principle behind MRI machines, so they would have been right had they made that wild guess). Third, science takes time and is a cumulative process requiring a few breakthroughs and many routine but important developments. The science underlying atomic clocks in GPS satellites took ~30-40 years to develop, another ~40 years to be incorporated into military technology, and another ~30 years after that to yield mass consumer products. It should also be noted that there are many important, unsung nuggets of progress along the journey of scientific discovery, and most big breakthroughs build on dozens of these. Thus, it is often not even useful or accurate to discuss science in terms of ‘breakthroughs’.
Ultimately, governments fund science because it yields applications and economic growth, but these applications are so far in the future and so unpredictable that the R&D cannot be privatized (more cynical reasons: pissing contests with other countries; keeping scientists employed so they don’t foment a revolution or mug people on the street in order to buy a telescope). Because science is a public good, akin to utilities and infrastructure, governments have a central role in fostering scientific innovations. Their main duty is to not eat their seed corn—to actually fund science in a reliable (don’t play politics with science funding), broad (don’t only fund a few fields or a few institutions), and sufficient (let’s stick to a first-world level of expenditure as a % of GDP) way, so that society may reap the benefits of an accidental discovery or a sustained decades-long effort in the future (as it is enjoying benefits of past investment today).
No one knows what the winning ideas will be, but individual scientists know what interests them (and often it is purely curiosity driven) and they (should) know what has not been explored before. Universities, where much of the modern scientific research takes place (government labs too), serve as the bridge between the individuals’ interests and the government’s/society’s interests. The main role of universities in fostering breakthroughs is to permit long-term sustained research and to create an environment where scientists can tempt serendipity. This has many factors including facilities, colleagues, administration, figuring out what to do with all the junior scientists for whom there are no permanent positions. But the main ingredient that universities can uniquely provide is stability: there are many institutions where short-term (<5 years) research takes place, but fewer institutions where someone can sustain a research program for decades, allowing for slow-and-steady progress, and a large cross section for serendipitous events/discoveries. Research universities should not lose sight of this.
Science is a social process, and journals serve as the gatekeepers of formal scientific communication. Ultimately, they are not too important—no one really remembers the names of the journals in which historically important articles were published. However, they can be important in the short term insomuch as they affect a scientist’s ability to get government funding or attain/sustain a position in a university or other research institution. The best thing journals can do is to reduce unnecessary friction in formal scientific communication, and in my opinion, the best way to do this is to move to a not-for-profit, open access model. The open-access part should be obvious, but not profiting from the content of a scientific journal is equally important for ensuring that the content is published due to its merit, not its tabloid appeal.
It is probably unsatisfying that my best suggestion is to do what we are currently trying to do but with more money and less hype. If one is seeking a more radical policy direction, my suggestion is to recreate Bell Labs (profit motive, government monopoly, manufacturing and all). Bell Labs was able to foster many of the most important research results of the 20th century, both applied (transistor) and basic (cosmic microwave background) (see: Inna Vishik’s answer to Why was Bell Labs in New Jersey able to do so many cool things before they split up?). There are many reasons for this scale of innovation, but a key factor is that they were trying to solve one of the most difficult problems of their day (how to connect the US and the entire world?) which happened to touch on many different areas of science. There are modern problems—alternative energy, transportation, climate change—which similarly have a long time horizon, bridge many research fields, and require government involvement, which may be well suited for another idea factory.
See also: What are some interesting cases where a piece of science (math, physics, etc.) was thought to be only theoretical but turned out to have great applications?
This question originally appeared on Quora – the knowledge sharing network where compelling questions are answered by people with unique insights. You can follow Quora on Twitter, Facebook, and Google+. More questions:
Scientific Research: What is the most frustrating thing about being a scientist?
Science: Where could a person feel the maximum gravity on Earth?
Colleges and Universities: Why do we often hear about perfect SAT scores, but never about perfect GRE scores?
Forbes
