A Response to Sabine Hossenfelder
This is a response to Sabine Hossenfelder’s video:
Scientific Progress is Slowing Down. But Why?
In it she asks whether science is dying.
She references research that seems to indicate a fall in scientific productivity.
Is Sabine right? Is science dying? Is scientific productivity declining?
This is a response, not an attack. I am not debunking her video — I wouldn’t dare!
But I am going to offer a different perspective.
But, first, what is it that scientists do?
So What is it that Scientist Do?
To my mind scientific research can broadly be divided into three categories:
- Exploration
- Grand Synthesis
- Exploitation
Let’s deal with each of them in turn.
Exploration
The purest example that comes to mind is Galileo pointing his telescope at the sky. I don’t know what he expected to find but I doubt it was the moons of Jupiter.
Maybe he did think he might find the phases of Venus or the mountains of the moon.
Other examples are Antonie van Leeuwenhoek making his microscopes and discovering tiny life forms invisible to the naked eye, Aristotle dissecting animals to see what was inside and William Crookes’ “Crookes tube” which led to the discovery of electrons. (It also led to the first TV sets using cathode ray tubes. Anyone remember them? Ever wonder how Youtube got its name?)
Modern examples include the James Webb Space Telescope and the LHC.
Another name for “exploration” is “data gathering”.
Grand Synthesis
Why “grand synthesis”? Why not just “synthesis”?
Grand synthesis is the process of coming up with some sort of overarching theory that explains a large variety of observations in terms of a few simple assumptions. Generally speaking, a grand synthesis changes the way we think about the whole world. It results in a “paradigm shift”.
The first name that springs to mind, the original “grand synthesist”, is Isaac Newton and classical mechanics.
Other “grand synthesists” include John Dalton (atomic theory), Charles Darwin and Alfred Russell Wallace (evolution by natural selection), Gregor Mendel (genetics), James Clerk Maxwell (equations of electromagnetism), Albert Einstein (special & general relativity, quantum mechanics), Erwin Schrödinger (the wave equation), Edward Lorenz (deterministic chaos) and, finally, the one you’ve never heard of, Peter Dennis Mitchell (how do organisms turn food into useable energy).
All of these changed the way we thought about the world.
Obviously the border between exploration and synthesis of any sort is fuzzy. Any experiment may be considered exploration since we don’t know what we’ll find. If we did, we wouldn’t bother doing the experiment.
However, many, perhaps most, experiments are not random exploration. They are performed to confirm or falsify some hypothesis. A modern example is using the LHC to find the Higgs boson.
Of course the most exciting experiments are those that turn up something completely unexpected. A fairly recent example is the discovery of nuclear fission.
In December 1938, at the Kaiser Wilhelm Institute for Chemistry in Berlin, Otto Hahn and Fritz Strassmann bombarded uranium salts with slow neutrons. To their surprise they discovered barium in the resultant solution. They transmitted these findings to Lisa Meitner in Stockholm whence she had fled in 1938.
Meitner and her nephew, Otto Frisch, provided the explanation. The neutrons had split the uranium nucleus. This was totally unexpected.
(See “The Discovery of Fission” by Otto Hahn, Scientific American, February 1958, pp76–84)
What is interesting about this is that it is impossible to pin down the discovery of fission to any one person. Hahn and Strassmann did the experiments. Meitner and Frisch provided the explanation. Only Hahn got a Nobel Prize. Bear this in mind in what follows.
Exploitation
This is where science blends into technology.
Think about the Manhattan Project. It did not involve any new scientific principles. There was no grand synthesis explaining many phenomena based on a few simple assumptions. There was no paradigm shift. It was purely an engineering problem. The only question to be answered:
Was it possible to create a nuclear weapon with the technology of the 1940s?
It was a huge collaborative effort. No one person on his own could have done it. It is impossible to point a finger at an individual and say, “He invented the A-bomb.”
Of course there are still individuals who come up with brilliant ideas. A recent example is the development of “ Bioorthogonal and Click Chemistry” by Carolyn R. Bertozzi, Morten P. Meldal and K. Barry Sharpless. The trio shared the 2022 Nobel Prize for Chemistry. (This was Sharpless’ second Nobel. He also shared the 2001 prize.)
Among other things, these techniques enable scientists to uncover some of the tricks cancer cells use to evade detection by the immune system.
There are two things to note about this work, important and ground breaking as it is
- Bioorthogonal and click chemistry does not involve some grand synthesis that changes our understanding of the whole world in the way that quantum mechanics or evolution by natural selection did. Instead it gives us a more detailed understanding of some important biological processes.
- All three of the Nobelists head laboratories that employ other researchers who did much of the work over a period of many years. In fact, near the beginning of her Nobel speech, Carolyn Bertozzi references the work done by “my co-workers and I”
Well, I’ve gone into far more detail than I’d planned. But what conclusions can we draw?
Some Conclusions
Exploration and exploitation are both flourishing.
Grand synthesis had a glorious three hundred-year run from Newton’s Principia to QFT and the Standard Model. Now it’s hit a wall. It’s been stagnant for about 50 years. There have been no grand new theories that have altered our understanding of the whole world.
With the benefit of hindsight we can say the completion of the Standard Model marked the end of an era of grand synthesis in science. It was heady stuff while it lasted. We all mourn its passing.
But, for now at least, it’s gone. In this sense Sabine is right. Grand synthesis is not merely dying. It’s at best moribund.
What about declining scientific productivity?
Sabine cites analyses of scientific papers that purport to measure scientific productivity.
But is this still a valid measure?
How do you measure scientific productivity anyway?
Is it even reasonable to measure it at the level of individual scientists?
We don’t measure the engineering prowess of SpaceX solely by examining Elon Musk’s personal activity. We recognise he assembled an outstanding team that did most of the development under his guidance.
Isn’t the same true, at least to some extent, of the three winners of the 2022 Chemistry Nobel Prize?
Consider the microchip industry. Thousands of people are currently working on developing a new generation of faster, more energy efficient chips. For a good summary of the state of play here’s a link to a short video.
New Disruptive Microchip Technology and The Secret Plan of Intel
Little of this work will appear in scientific journals. No one will be able to point a finger at some individual and say she or he invented a new generation of microchips. It will not in and of itself result in any mind-changing new paradigms.
But let’s suppose these efforts do result in a new generation of chips. Are we going to say that an effort which occupied the minds of some of the smartest scientists on the planet for many years was not “progress” because it did not result in papers in scientific journals.
Are papers really a good measure of scientific progress?
Final Thoughts
I don’t know how you measure scientific productivity so I can’t say whether, or not, it’s declining.
I don’t think science is dying but I do think since the completion of the standard model we’ve entered a different era in which the old measures of progress no longer apply. papers in scientific journals are no longer the be all and end all of science. Perhaps they never really were. Perhaps they only ever applied to certain aspects of the scientific enterprise.