How to Escape Scientific Stagnation

IMHO, ARPA-H and other new agencies need to change how funding is allocated.

Right now, in experimental sciences, lots of professors from prominent universities have a monopolistic attitude towards research. They have just become middlemen. They manage large groups, have become completely detached from technical things and use their position to secure most funds in the field and also veto others from publishing or getting funded. This is a recipe for disaster, stagnation and fiascos as we can see e.g. in the field of Alzheimer's disease.

Innovative laboratories like LMB Cambridge or Cold Spring Harbor, where a great deal of work was done during the golden age of molecular biology, were at odds with this approach. All funding was internal, and the number of people managed by senior researchers was capped to a single digit.

In short, we need funding policies that weed out scammers and promote good technical work. ARPA for example is quite aware of this problem, and IMHO has much better policies than, say, the European Research Council (ERC). ERC is quite infamous as they don't even bother about withdrawing funds from laboratories that are known to have manipulated research results and had to retract their publications (produced using ERC's funding!). https://retractionwatch.com and https://forbetterscience.com are full of these stories.

I sometimes think about how eccentric the leading scientists of previous generations were. Kurt Gödel for instance would only eat food prepared by his wife, and starved to death when she fell ill and couldn’t cook for him. Just look at a picture of Albert Einstein and you see he was a bit of a character. To say the same about Richard Feynman you may need video, but it’s quickly quite clear.

When you consider how “political” everything has become over the last few decades it seems likely to me that this generation’s Gödels, Einsteins and Feynmans where “outmaneuvered” early in their careers by more socially adept rivals and never reached any positions of influence. Sadly I think you see this pattern in any high status field. The few that love the nitty gritty and can really contribute are typically outmaneuvered by the many who are drawn to the prestige.

Measuring scientific progress is tricky. The elucidation of the structure of DNA transformed genetics and biology. But was it bigger than recombinant DNA, or CRISPR? In some ways yes, because we knew so little before that breakthrough. But we keep making unexpected and even sometimes revolutionary discoveries. It is a lot easier to make big discoveries when little is known. So, on a percentage basis, the more you know, the harder it is to have a big impact. But recombinant DNA, embryonic stem cells, CRISPR, were transforming discoveries. Molecular biology was not "done" once the central dogma was established.

"In a working paper published last year, Chiara Franzoni of the polimi Graduate School of Management and Paula Stephan of Georgia State University look at a number of measures of risk, based on analyses of text and the variability of citations. These suggest science’s reward structure discourages academics from taking chances"

I believe the focus should be in making more basic science affordable to private parties where the incentives for innovation are more clear. Scientific instruments and equipment are incredible expensive, taking action to make them affordable and within reach of more startup companies would definitely have an impact.

Easy: force all tenured professors and researchers to retire at 68

This is exactly the mindset that holds science back. While we often romanticize the idea of the eccentric genius who has some brilliant eureka moment and revolutionizes a field, such instances are an incredibly small minority of scientific progress. Every once and again there is indeed a paradigm shift that opens up a new area of knowledge to explore, but then there are decades of methodical effort actually exploring that space. A small improvement in one area can be used to make a more precise measurement in another, which in turn gives further insight to some third thing. And ultimately, while we might not give out nobel prizes for something as "mundane" as a mild refinement to semiconductor manufacturing or drug synthesis, the value to society of any such development might be many billions of dollars.

High risk, high reward moonshots might sound like an appealing alternative to the current strategy of methodical investigation, after all it's just a numbers game and a you only need a few wins to justify many attempts. But unfortunately it does more harm than good. A single experiment without the context to interpret the data it yields or the commitment to refine its operation is not going to lead to new understanding. Unless you were incredibly lucky and just happened to set everything up right to strike gold on the first try, odds are you will see nothing and convince people that therefore there is nothing of interest to be gained by continued pursuit. In analogy to actually prospecting, you're much more likely to hit paydirt by taking the time to study an area's geology and identifying good spots to dig than digging holes in the ground at random.

Finally, the very premise that scientific stagnation is a problem we need to deal with is shaky. While it is very difficult to quantify scientific progress, various proxies like numbers of papers published, numbers of subdisciplines spun off, number of patents filed, etc all seem to indicate that progress not only is ongoing but has been steadily accelerating. While the change in our understanding of the world from any one discovery might be less and less, there are so many more possible avenues for discovery than at any other time in history, and each discovery expands our frontiers to expose new opportunities. There is no shortage of scientific inquiry and ideas, the only limit is that we have advanced so far that it is becoming hard to find natural phenomena at measurable scales that our models of the world do not already fully explain. We found all the particles of the standard model, we can construct things on atomic scales, genetic engineering has become mundane and protein folding is solved, we have mapped the earth from space with millimeter precision and are in the process of doing the same to other worlds, we can accurately simulate airflows at hypersonic speeds and the cores of collapsing stars. Most of these achievements would have been gibberish a mere century ago, and the achievements yet to come will be just as unimaginable to us if not more so.

Paywall!

It's actually capitalism that has stagnated scientific progress. Initially capitalism revolutionizes production by liberating it from the backwardness of feudalism, establishing division of labor, science over church, automation, etc.

But after a long process of development, capitalism becomes a fetter on scientific discovery. It requires a permanent underclass which is never able to develop their talents. So most humans are trapped doing repetitive and mind-numbing work. Like Stephen Jay Gould said of Einstein there's a: "near certainty that people of equal talent have lived and died in cotton fields and sweatshops"

Also capitalism is generally focused on short term profit rather than projecting long term social needs. This is a major reason we're sprinting toward climate catastrophe when the technology exists to overcome it. Capitalism cannot put the resources and manpower available to work solving the glaring needs of our time. This is the main reason there is scientific (also cultural) stagnation.